Yutian Lei
Papers
3
Total Citations
37
H-Index
2
About
Yutian Lei is a rising star in robotics and artificial intelligence, whose research focuses on bridging the gap between large language models (LLMs) and sample-efficient robotic manipulation. Lei’s major contributions lie in developing frameworks that leverage the reasoning and internal knowledge of LLMs to dramatically improve reinforcement learning (RL) in real-world robotic tasks. In their highly cited work, "RLingua," Lei proposed a novel framework that uses LLMs to reduce the sample complexity of RL for robotic manipulations, addressing one of the field's most persistent bottlenecks—earning 23 citations since 2024. Their follow-up, "RT-Grasp," extends this by enabling multi-modal LLMs to perform reasoning-tuned robotic grasping, moving beyond text-based planning to direct action generation. Lei also introduced the Virtual In-Hand Eye Transformer (VIHE), a method that enhances 3D manipulation through action-aware view rendering, allowing for multi-stage action refinement. With a total of 37 citations across these recent papers, Lei’s work is quickly gaining traction for its practical, data-efficient approach to teaching robots complex tasks. Their research is particularly notable for demonstrating how LLMs can serve as cognitive scaffolds for RL, making it a must-read for anyone interested in the future of intelligent robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3VIHE: Virtual In-Hand Eye Transformer for 3D Robotic Manipulation2 citations · 2024